[RockvilleBob]

The Auto Focus section of the Pentax review seems to have a few areas that concern me. As I have stated before comparing the performance of the K-3 with the Pentax 60-250mm F4 to the Nikon D610 (w/ Nikon 70-200mm 2.8) is like comparing a car that costs $31,000, for example a Nissan Maxima, to the Chevy Corvette costing $51,000. Both the Nikon camera and the Nikon lens are far more costly. It is not an apples to apples comparison.

But there are other sections which seem to go out of their way to shine a false poor light on the K-3. For example take the statement: "The expanded area AF mode (SEL 4) often "expands" to an entirely different subject, which requires the photographer to let go of the shutter button and re-focus once he or she realizes the error, which may not be easy when your subject is far away." Why was such a small area used if only to show that the subject could leave the area? This is the first mention of SEL4 -the test described was done using the expanded area SEL-27. The only reason I can come up with is to be able to say camera lost focus but waht about the subject and the photographer? Was the photographer able to keep the subject in the SEL 4 area - to me it doesn't appear like he or she did - so blame the camera instead of using the SEL 27 area. Why did the reviewer use SEL 4 to highlight this occurrence? What happens when a photographer using a Nikon camera chooses to reduce the focus points/area used by 85% so instead of using the 39 points only 5 or 6 points would be used?

Then there is the statement: "We'd like to note that the autofocus performance you experience with the K-3 will greatly depend on the lens that you're using. A modern DC/SDM (silent drive) lens, such as the 18-135mm, focuses very quickly. On the other hand, screwdrive lenses with long throws, such as the DA 55-300mm, still feel like equipment from the last decade, even when paired with the K-3. Primes with shorter focus throws such as the DA limiteds will focus quickly, but adjustments to the AF are nowhere near as smooth as with DC/SDM lenses." But where is the lens focus speed comparison of the Pentax 60-250mm F4 versus the lens focus speed of the Nikon 70-200mm 2.8. I really think my DA*300 is great but it glides into focus it certainly doesn't snap into focus. Is the performance of the Pentax Pentax 60-250mm F4 similar (which I do not own?) What about the focus speed of the Nikon 70-200mm 2.8, another lens I do not own. I will hazzard a guess here that the Nikon 70-200mm 2.8 is a lens that snaps rather than glides into focus.

So the review makes the point that lens plays a big part in AF performance yet when comparing the Penatx camera with one lens to a Nikon camera with another lens there is no mention of the relative lens focusing performance. I will hazzard another guess here - the lens performance between the two is at least as large as the 100% price premium of the Nikon over the Pentax.

Why editiorial comment: In terms of comparing the Pentax to the competition I feel the author of the auto focus section went the extra mile to find areas to fault.

[Adam]

This is the first mention of SEL4 -the test described was done using the expanded area SEL-27

SEL-4 uses all 27 points, SEL-3 uses 25, SEL-2 uses 9, SEL-1 uses 1 as per the terminology on the camera's top/vf LCD. These settings are also referred to as SEL-L, SEL-M, SEL-S, and SEL on the info screen, respectively.

The comparison done in the review is to put things into perspective, nothing more. The only conclusion being drawn is Pentax hasn't revolutionized its autofocus performance with the K-3. We took two top-shelf and comparable lenses from each manufacturer, which is a very realistic test IMO. The D7100 may have been fairer to use as the competing body, but a. people who are considering the K-3 are probably also just as interested in the D610 as they are in the D7100 and b. the D610 has an old AF system, so the D7100 would do better if anything.

[StigVidar]

Your test of K-3 againt D610 is OK by me :-)

But I was wondering about this "The expanded area AF mode (SEL 4) often "expands" to an entirely different subject, which requires the photographer to let go of the shutter button and re-focus once he or she realizes the error, which may not be easy when your subject is far away." statement, too.

Did it happened in the test against D610? Would you get more sharp pictures if this was not the case?

[spartan]

Adam

Would it be too much to ask to rerun the test with the fastest AF lens from Pentax, which may be the 18-135?

[woodywesty]

Wouldn't a better test comparison have been using the NIKKOR AF-S 70-200mm f/4G ED VR Telephoto which is the same price as the Pentax zoom used in the test??
In any event, I think that always comparing Pentax lenses and camera bodies with Nikon and Canon is over done. Pentax is and has always produces quality camera gear which in many instances is as good as any gear. The biggest difference is that Canon and Nikon just offer more options. If Pentax/Ricoh chooses to produce $2,000 and $3,000 camera bodies and $8,000 lenses then they will be every bit as good as anything produced by Canon and Nikon, but I doubt that they will ever do that as the market for the very high end gear is pretty darn small....

[DDoram]

It is very difficult to test AF systems so any test will be flawed. This particular test has many flaws as has been pointed out but as a "system" comparison, it is of interest. However, since the review can out within a week of receiving the camera, I would suggest that it will take longer than that to get familiar with the various AF settings and how they affect AFC performance.

Dale

[Kunzite]

The comparison done in the review is to put things into perspective, nothing more. The only conclusion being drawn is Pentax hasn't revolutionized its autofocus performance with the K-3.

As a software engineer, I'm painfully aware how important it is to know exactly what you're testing for.

If you want to see if Pentax has "revolutionized" its autofocus performance, you have to compare exactly that - old body vs new body, all other things being constant. You've got advanced features like the Real Time Scene Analysis system (left untested), an increase in AF points count with no performance penalty and, in your test, some 60% increase in keepers count.

If you want to see if Pentax K-3 with a 60-250 can keep up with a D610 with a 70-200 ($2700 vs $4400 setups), then you have to compare them. However, this won't tell you if Pentax "revolutionized" or not its autofocus.

I'm afraid you mixed two separate subjects, and drawn an incorrect conclusion.

[RockvilleBob]

The review in the forum is quoted at other photography areas on the internet. For example this one:
Pentax K3 indepth review @ Pentaxfrums - Steve's Digicams Forums
Where it flat out says Pentax auto focus lags behind Nikon. It does not say that the test was performed using a system from Nikon where just the lens costs about the same as the Pentax Camera and lens, it doesn't say that the two systems compared are vastly different in price. It just says that Pentax AF lags Nikon AF based upon the review in the Pentax Forum.
The review did not properly test one camera's auto focus versus another.
I still feel the existing review in the AF section as performed and as written is a disservice.
am venturing a guess that if the Nikon 7100 was used and the AF test as described were performed using RAW images then the Nikon would rank below the Pentax untis in the test, even with the spiffy $2300 lens. If that result were the case would it be proper to say that Nikon AF lagged Pentax - absolutely not.
I feel the test can only conclude that the Pentax system as tested performs below the Nikon system as tested.
The conclusions drawn in this test are being interpreted incorrectly because the test was improper in its format except to test the Pentax K-3 equipped with the 60-250 versus the Nikon D610 equipped with the 70-200.
As I stated elsewhere I feel the forum review and test of auto focus should be rewritten or expanded in the auto-focus area to compare apples vs apples.
I am guessing if the Nikon 7100 were used in a test using a common lens - thereby comparing one camera versus another - the K-3 would outperform the Nikon, probably by a very wide margin in raw.
The existing review gives people looking at making camera decisions a incorrect comparison. The only valid comparison is for a person who is going to decide should I buy a Pentax K-3 with a Pentax 60-250 or should I buy the Nikon D610 with the Nikon 70-200 f2.8. To me that is like deciding if I should by a Nissan Altima or a Chevrolet Corvette. If you want a Corvette you should buy it if you can afford the price. If you don;'t have that much money to spend on another car then a comparison of a Corvette against an Altima provides information of little relevance.
The intent of the test was good, the execution was probably done quickly to get the review published, the conclusion is not warranted by this test.

[Rondec]

It has been said by many people that SDM lenses need to be updated. That is true, without a doubt, but all that this test shows is that SDM is a real drag on the auto focus performance. I know that. I have a hard time tracking anything with my DA *50-135 or 55, but the question is if the K3 has improved tracking with other lenses (like Sigma HSM lenses, or screw driven lenses) the answer to that question does seem to be yes.

[Adam]

If you want to see if Pentax has "revolutionized" its autofocus performance, you have to compare exactly that - old body vs new body, all other things being constant.

There was a K-5 IIs comparison in there too. We did just that!

but the question is if the K3 has improved tracking with other lenses

IMHO, yes, there is a slight improvement because of the higher-density AF point layout. But it's not night and day.

[nomadkng]

Therefore the only test that could be made would be using a lens available to both systems?

So get a same vintage sigma 50-500 (which is a popular wildlife/sports lens) and test the 2 cameras against each other that way?
Both screw drive and HSM?
One should also run comparisons against the K5/K5II and both the 610 and 7100 for imperical conclusions?

That's a lot of testing and data collection, but seems to be the only answer?

[Kunzite]

There was a K-5 IIs comparison in there too. We did just that!

IMHO, yes, there is a slight improvement because of the higher-density AF point layout. But it's not night and day.

You surely did (though it would have been nice to know the test conditions and results in more details). The effort of trying to cover so much in a review - one that was quickly published - is commendable. Of course you can't do everything, and perfectly so.

What I'm saying is that your conclusion, that "Pentax hasn't revolutionized its autofocus performance with the K-3", should be based solely on the comparison with the K-5II, and not with the D610.
And from your review, we're seeing a significant increase in keeper count (60%), and we have a completely new (for Pentax) territory to explore - linking AF with RGB metering is opening new possibilities for Pentax. Because raw speed is one aspect of performance, the other is being able to properly select and follow the subject (and the K-5II is pretty much "dumb" in this matter).

[azimuthal]

The review in the forum is quoted at other photography areas on the internet. For example this one:
Pentax K3 indepth review @ Pentaxfrums - Steve's Digicams Forums
Where it flat out says Pentax auto focus lags behind Nikon. It does not say that the test was performed using a system from Nikon where just the lens costs about the same as the Pentax Camera and lens, it doesn't say that the two systems compared are vastly different in price. It just says that Pentax AF lags Nikon AF based upon the review in the Pentax Forum.
The review did not properly test one camera's auto focus versus another.
I still feel the existing review in the AF section as performed and as written is a disservice.
am venturing a guess that if the Nikon 7100 was used and the AF test as described were performed using RAW images then the Nikon would rank below the Pentax untis in the test, even with the spiffy $2300 lens. If that result were the case would it be proper to say that Nikon AF lagged Pentax - absolutely not.
I feel the test can only conclude that the Pentax system as tested performs below the Nikon system as tested.
The conclusions drawn in this test are being interpreted incorrectly because the test was improper in its format except to test the Pentax K-3 equipped with the 60-250 versus the Nikon D610 equipped with the 70-200.
As I stated elsewhere I feel the forum review and test of auto focus should be rewritten or expanded in the auto-focus area to compare apples vs apples.
I am guessing if the Nikon 7100 were used in a test using a common lens - thereby comparing one camera versus another - the K-3 would outperform the Nikon, probably by a very wide margin in raw.
The existing review gives people looking at making camera decisions a incorrect comparison. The only valid comparison is for a person who is going to decide should I buy a Pentax K-3 with a Pentax 60-250 or should I buy the Nikon D610 with the Nikon 70-200 f2.8. To me that is like deciding if I should by a Nissan Altima or a Chevrolet Corvette. If you want a Corvette you should buy it if you can afford the price. If you don;'t have that much money to spend on another car then a comparison of a Corvette against an Altima provides information of little relevance.
The intent of the test was good, the execution was probably done quickly to get the review published, the conclusion is not warranted by this test.

the car analogy has limited usefulness, buying a car doesn't limit you to buying cars from that same manufacturer later, or even incentivize it.

the review just shows that top of the line nikon still holds an edge over top of the line pentax in the realm of autofocus. the review was more useful than the early user reports of the pentax k[insert number] where "pentax miraculously revolutionized autofocus and it's just as good [or better] than everyone else".

the review is less useful to you if you want to pick a specific budget and then compare merits that way. but just because pentax doesn't make a "corvette" doesn't mean you can't appreciate them.

if money is not an issue, and autofocus is your biggest concern, other systems have been better in this sole regard, that has been a constant since pentax made digital SLRS, it doesn't lessen the other benefits of choosing pentax over something else, and it hasn't prevented people from taking great pictures with pentax systems.

[bxf]

As a software engineer, I'm painfully aware how important it is to know exactly what you're testing for.

If you want to see if Pentax has "revolutionized" its autofocus performance, you have to compare exactly that - old body vs new body, all other things being constant. You've got advanced features like the Real Time Scene Analysis system (left untested), an increase in AF points count with no performance penalty and, in your test, some 60% increase in keepers count.

If you want to see if Pentax K-3 with a 60-250 can keep up with a D610 with a 70-200 ($2700 vs $4400 setups), then you have to compare them. However, this won't tell you if Pentax "revolutionized" or not its autofocus.

I'm afraid you mixed two separate subjects, and drawn an incorrect conclusion.

My thoughts precisely, but I'd take it a couple of steps further:

- which system gives me the better AF for a certain amount of money?
- Using comparable components, who has the better AF? (i.e. K-3 vs D7100, both using the same 3rd party lens)

[RockvilleBob]

the car analogy has limited usefulness, buying a car doesn't limit you to buying cars from that same manufacturer later, or even incentivize it.

the review just shows that top of the line nikon still holds an edge over top of the line pentax in the realm of autofocus. the review was more useful than the early user reports of the pentax k[insert number] where "pentax miraculously revolutionized autofocus and it's just as good [or better] than everyone else".

the review is less useful to you if you want to pick a specific budget and then compare merits that way. but just because pentax doesn't make a "corvette" doesn't mean you can't appreciate them.

if money is not an issue, and autofocus is your biggest concern, other systems have been better in this sole regard, that has been a constant since pentax made digital SLRS, it doesn't lessen the other benefits of choosing pentax over something else, and it hasn't prevented people from taking great pictures with pentax systems.

My guess is straight up the K-3 will outperform the Nikon 7100 in raw continuous focus testing. That does not entitle one to say Pentax AF is better than Nikon. I will hazard a guess that the Pentax K-3 is the best AF cropped camera currently made. Reading the review one would never come close to guessing that might be possible. I feel the test was a case of self fore-filling prophecy designed to prove the testers initial assessment.